Deep Learning and Computer Vision in R: A Practical Introduction

BIBC2025 workshop

Patrick Li

RSFAS, ANU

About me

I am Patrick Li

  • Education background
  • Current work
  • Past experience in CV

Content summary

  • Overview of computer vision (CV)
  • reticulate basics
  • Neural networks
  • Image classification
  • Hyperparameter tuning
  • CV model interpretation
  • Object detection
  • Image segmentation

Disclaimer

Code and theory

This workshop is a hands-on introduction to modern computer vision in R. You’ll work through code examples and practical exercises. Theory will be covered only to the extent needed to understand the key ideas and statistical foundations behind the models.

Reproducibility

Computer vision models can be only partly reproducible due to GPU computation. Don’t be surprised if your results differ from those shown in the materials.

Expectation


How confident should I be coming into this workshop?


If you have a statistics background, you are good to go!

There will be some new concepts and terminology, but nearly everything can be explained in statistical terms. Don’t be intimidated by CV jargon.

Overview of computer vision

Computer vision

Computer vision is a broad field concerned with enabling machines to interpret and understand visual information from the world.

  • Image classification: Assign a single label to an entire image.
  • Object detection: Locate and classify multiple objects within an image using bounding boxes.
  • Image segmentation: Label each pixel with a class, producing a detailed mask of objects.
  • Image generation: Create realistic or artistic images from scratch or from a condition.
  • Action recognition: Identify actions or activities in images or video sequences.
  • Visual Question Answering: Answer questions about an image by understanding its content.